Data mining using a genetic algorithm‐trained neural network

Abstract

Neural networks have been shown to perform well for mapping unknown functions from historical data in many business areas, such as accounting, finance, and management. Although there have been many successful applications of neural networks in business, additional information about the networks is still lacking, specifically, determination of inputs that are relevant to the neural network model. It is apparent that by knowing which inputs are actually contributing to model prediction a researcher has gained additional knowledge about the problem itself. This can lead to a parsimonious neural network architecture, better generalization for out‐of‐sample prediction, and, probably the most important, a better understanding of the problem. It is shown in this paper that by using a modified genetic algorithm for neural network training, relevant inputs can be determined while simultaneously searching for a global solution. Copyright © 2001 John Wiley & Sons, Ltd.

Novel Approaches to Probabilistic Neural Networks Through Bagging and Evolutionary Estimating of Prior Probabilities

“…The best generalization accuracy ever for the Heart problem has been achieved by a MLP that its weights and connections (architecture) were commonly evolved by a Genetic Algorithm [22]. Finally, for the Horse dataset, the best result has been achieved by a NN trained by a Genetic Algorithm and a feature selection procedure [27]. As we can see, the performance of the proposed approach is more or less close to the best model's performance on each dataset, besides the Glass where there is a great superiority of the SMC-RBF network.…”

Optimasi Nilai Bobot Algoritma Backpropagation Neural Network Dengan Algoritma Genetika

“…Algoritma Genetika akan digunakan untuk menentukan bobot serta bias awal terhadap parameter Backpropagation sehingga bisa mendapatkan kemampuan belajar yang baik. Menurut (Sexton and Sikander 2001) Algoritma Genetika memulai dengan multiple random point sebagai inisial populasi ketika mencari solusi. Setiap solusi kemudian dievaluasi berdasarkan fungsi objektif, setelah selesai solusi tersebut kemudian dipilih untuk generasi selanjutnya berdasarkan fitness mereka.…”

Using Evolutionary Algorithms for the Scheduling of Aircrew on Airborne Early Warning and Control System

“…Genetic algorithms are state-of-the art methods that aim to optimise functions by modeling the nature 26 . It was first introduced by John Holland 27 and continues to be developed by many researchers today [28][29][30][31][32][33][34][35][36][37][38][39][40] . Genetic Algorithm parameters represent the genes of chromosomes, while all parameters form chromosomes.…”